Incident analysis of Paris RATP metro lines
Project description
Statistical analysis of incident probability and causes on RATP metro lines
Trafic perturbé...again?! This library and this notebook provides a statistical point of view for all these daily incidents occurred on the Paris RATP metro lines.
Using tweets coming from official RATP Twitter accounts (@Ligne1_RATP for line 1 for example), we will see
- What is the probability of encountering some operational incidents on a particular line?
- Which lines are more likely to cause everybody unhappy?
- What are the main causes of these problems?
- Should I be considered lucky if I never take metros during rush hours?
- Are there less problems during weekends?
- Instead of going on vacation, is there any reason to be happy if I still work in August?
Some examples in the year 2018
The fist two examples below refer to the RATP metro line 2, which I take everyday for work.
I should leave work before 17:00 on Wednesday (or after 21:00)
In the figure below you can see the probability of catching an operational incident (trafic perturbé, interrompu...) at a given hour (in fact in the next following hour) and a specific weekday. The maximum value (nearly 9%) can be found on Wednesday at the evening rush hours.
It's us the main responsible for these incidents
Nearly half (48%) of the incidents come from us. Also, 9% is due to unattended bags...
Line 13 should be jealous of line 4
Every Parisian knows that the line 13 is bad. But in the year 2018 it was beaten by the line 4, which records an average probability of incidents larger than 4%.
License
The code is licensed under the terms of the MIT license. The analysis results (text, figures, tables) are licensed under a Creative Commons Attribution 4.0 International License.
Author
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for ratpmetro-0.0.2-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | bd69defccb3d97fd8249f80e14a80e106d79c3c6db8e9b9e66aadc05e5bcec8d |
|
MD5 | 96bfbef5d98a24975df0aa5e9a834d1a |
|
BLAKE2b-256 | 847d8cc57fdebd3352b4e993e62c60b55d5e93204c590ab998c195110fad7164 |